A single index that can take in everything from bonds and equities to real estate and commodities from around the world could be a powerful investment tool © FT montage; Bloomberg; Reuters

Financial benchmarking giant MSCI is working on what some in the industry would dub an “ultimate index” tracking the performance of all markets, which could mark the culmination of half a century of academic theory and practical financial engineering*.

Indexing is a booming business, slicing markets up into geographies or categories such as equities or bonds, and then subdividing further by size or industry. These are then used as benchmarks for fund managers, or packaged up into investable products. But a way to combine everything from commodities to venture capital in one gauge has proven elusive.

“It’s the holy grail,” says Mark Makepeace, the head of index provider Wilshire. “It solves a core problem of investing and would be hugely beneficial.”

When Makepeace led FTSE Russell, the group teamed up with Nobel laureate and Stanford economics professor William Sharpe to create an adaptive asset-allocation tool that started to tackle the idea of an ultimate index, but included only stocks and bonds. 

A true ultimate index would go further, including other big asset classes, like commodities, and even private assets that do not trade on an exchange, like real estate, infrastructure, bank loans and stakes in hot Silicon Valley companies.

Peter Shepard, head of analytics research and product development at MSCI — one of the indexing industry’s Big Three, alongside FTSE Russell and S&P Dow Jones Indices — has been working on the quest to produce an index of this kind for more than a decade. He thinks he is getting close, although he is reluctant to term the final result an “ultimate index”.

Before he joined MSCI in 2007, Shepard was a theoretical physicist at Berkeley — his PhD thesis was titled On non-perturbative quantum gravity: Holography and matrix models in string theory — and he spent much of his time chasing the holy grail of physicists, a unifying Theory of Everything. He is therefore wary of applying the same label to the realm of indexing. 

Bar chart of Allocations of US public colleges, universities and foundations (%) showing How US endowment investment officers divvy up their money

“I need to be careful about not falling into that trap again,” he said. “I think we may be able to find a grand unified theory of physics at some point, but we’re not going to be able to find a grand unified theory of markets.”

The standard investment industry benchmark has long been the 60/40 portfolio — 60 per cent stocks and 40 per cent bonds. Vanguard’s US 60/40 fund has returned 168 per cent over the past two decades. Yet what to include in a broader index is hotly debated, and how to construct it is fraught with practical complications in sourcing reliable up-to-date data in areas like private markets.

Peter Shepard, head of analytics research and product development at MSCI

Despite that, Shepard is convinced they must be included. Private equity has returned about 12 per cent a year on average over the past 15 years, and private debt about 8 per cent, according to Morgan Stanley. “If you leave private assets out of this Platonic ideal, you’ll be leaving a whole lot of performance behind,” Shepard said. 

Even if one can come up with a decent mix of asset classes, figuring out how to sort by geography and other factors is not straightforward.

What is the best mix of US, European, Chinese or Brazilian stocks? How much exposure should there be to larger stocks or smaller ones? Should the bond allocation be weighted by the volume of debt issuance — as is standard for fixed income benchmarks — or is there a smarter way? The optimal answers are still uncertain, Shepard admits, but “simplicity and transparency are key”. 

“I could come up with a great black box, but if you don’t understand it, you won’t trust it and you won’t use it,” he said. “My hypothesis is that of the many ways that a chief investment officer adds value, some can be standardised.” 

The prospect is not just an object of geeky fascination in the indexing industry. If well constructed it could form the basis of cheap but powerful investment products for everyone from retirees to sovereign wealth funds, by simplifying the often arduous and expensive task of splitting money between different markets.

While the cost of investing in individual asset classes has been hammered down thanks to the invention of index funds, deciding on how to mix them is often handed over to a pricey financial adviser, or in the case of a pension plan, to a team of expensive professionals.

Bar chart of AQR's 5-10 year inflation-adjusted return estimates (%) showing Longer-term investment outlook for various asset classes

If one could assemble one broad investable benchmark for all assets — a true reflection of what Sharpe termed “the market portfolio” back in the 1960s, rather than a messy or facile proxy — then one could perhaps create a single, simple financial product suitable for most investors. 

“I think there’s a huge opportunity here,” Shepard said. “The asset allocation decision is the most important decision for a lot of investors . . . But we’re leaving this really critical decision to people who may not be that skilled, like my parents, or they turn it over to someone who charges fees for it.”

Despite the considerable hurdles, Shepard is optimistic that the quest for the Ultimate Index will soon bear fruit. However, he thinks that in practice the final result will be different flavours of a broad multi-asset benchmark.

“It might be a grand unified theory in terms of the framework. But customisation will be essential, as one size will not fit all investors,” he says. “At the same time, it has to be simple.”

*This article has been amended to clarify that MSCI has not itself labelled the project an “ultimate index”.

Email: robin.wigglesworth@ft.com

Twitter: @robinwigg

Get alerts on Asset allocation when a new story is published

Copyright The Financial Times Limited 2021. All rights reserved.
Reuse this content (opens in new window) CommentsJump to comments section